[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160615140028.GD32588@pd.tnic>
Date: Wed, 15 Jun 2016 16:00:28 +0200
From: Borislav Petkov <bp@...en8.de>
To: chenyu <yu.chen.surf@...il.com>
Cc: lkml <linux-kernel@...r.kernel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Thomas Renninger <trenn@...e.com>,
Kan Liang <kan.liang@...el.com>,
"Peter Zijlstra (Intel)" <peterz@...radead.org>,
"Rafael J. Wysocki" <rafael.j.wysocki@...el.com>,
Len Brown <len.brown@...el.com>,
Linux PM list <linux-pm@...r.kernel.org>
Subject: Re: [RFC PATCH] x86: Move away from /dev/cpu/*/msr
On Wed, Jun 15, 2016 at 12:36:22PM +0200, Borislav Petkov wrote:
> Ah, yes, good point. Will change.
Ok, got a bit more involved - I also need to disable preemption around
smp_call_function_single(). Here's what I have now. It seems to work
too.
static ssize_t
energy_policy_pref_hint_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
u64 epb;
preempt_disable();
if (rdmsrl_on_cpu(dev->id, MSR_IA32_ENERGY_PERF_BIAS, &epb))
epb = -1;
preempt_enable();
return sprintf(buf, "%d\n", (unsigned int)(epb & 0xFULL));
}
static ssize_t
energy_policy_pref_hint_store(struct device *dev, struct device_attribute *attr,
const char *buf, size_t count)
{
u32 val;
u64 epb;
if (kstrtou32(buf, 10, &val) < 0)
return -EINVAL;
if (val > 15)
return -EINVAL;
preempt_disable();
if (rdmsrl_on_cpu(dev->id, MSR_IA32_ENERGY_PERF_BIAS, &epb)) {
count = -EINVAL;
goto out;
}
if ((epb & 0xf) == val)
goto out;
if (wrmsrl_on_cpu(dev->id, MSR_IA32_ENERGY_PERF_BIAS, (epb & ~0xFULL) | val))
count = -EINVAL;
out:
preempt_enable();
return count;
}
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
Powered by blists - more mailing lists